DRAFT – PLEASE DO NOT DISTRIBUTE

Posture, Movement and the Alexander Technique

Dr Tim Cacciatore*

*With invaluable critique and input from Dr Patrick Johnson © Tim Cacciatore 2019

DRAFT – PLEASE DO NOT DISTRIBUTE

WHAT IS POSTURE? Everyone has a notion of what posture is, yet there's lots of controversy and confusion surrounding it, especially around how it relates to the Alexander Technique (AT). In fact, teachers’ opinions range from thinking that the AT is primarily about posture to thinking it’s not about posture at all! We’re not alone here – there’s plenty of confusion outside the AT community, even amongst clinicians and scientists. The phenomenon of posture is more elusive than it seems. In our experience a rigorous understanding of posture can help to think about the AT, both theoretically and practically. Posture is a much broader term than many AT teachers realise. The aim of this article is to clarify what posture is, how it’s controlled by the brain and how it differs from movement. In so doing, I hope to show that a modern scientific concept of posture is extremely relevant for AT teachers. First let’s look at a non-technical definition of the word. The Oxford Dictionary defines posture as “the position in which someone holds their body when standing or sitting”, in other words the relative positions of the component segments. Moreover, the word “posture” implies that the described body configuration is sustained – you wouldn’t refer to joints in motion as having a posture. Colloquially, when people refer to someone’s posture they often specifically mean spinal shape, typically with the notion that a straighter spine is better. In a nutshell, in common usage posture is synonymous with body position, sometimes with the view that some positions are better than others. At first glance, this definition doesn’t seem pertinent for the AT. We don’t typically teach one to “stand up straight” or “hold” a particular position. The positional shifts that occur result indirectly from changes in things like tension, attention or intention. Moreover, the AT has broad effects on motor actions such as movement, balance and breathing. In this context position seems trivial – isn’t the AT about a vastly broader range of habits than just position? Before we give up on the word posture, we need to recognise that the common definition does not tell the whole story. A more scientific perspective reveals posture as a much deeper concept than just “holding a position”. The common definition of posture neglects what’s going on physiologically to maintain the position. Yet it’s precisely this “postural maintaining” system that you must interact with to change carriage. One of my teachers described the AT as concerning postural behaviour, with the emphasis on behaviour to indicate that it’s the intricacies of how posture is maintained – not the specific position – that matters. Understanding the science of postural behaviour shows us that in fact posture is fundamental to our work.

DRAFT – PLEASE DO NOT DISTRIBUTE Eye as a model systemA To gain some insight into how posture is maintained physiologically, it’s useful to start simple. Believe it or not, the eyes are an ideal model system (yes, the eyes have posture). The brain circuits that control the eye are among the most studied and best understood circuits in the brain. Researchers now understand the brain circuitry that controls the eye and how it works and can therefore describe the eye’s postural behaviour precisely. Eye behaviour can be divided into two main modes: stationary phases when gaze is fixed on a point in space, and intermittent jumps called saccadesB. The stationary, or gaze holding, phases are the eye’s posture — it maintains the position in the socket. By contrast, during saccades the eyes rotate extremely quickly, jumping between gaze positions in 1/5th to 1/50th of a second. Surprisingly, you don’t see during saccades as the eyes are moving too quickly to produce an image. It is only during the gaze hold periods that you see, and even here any drift in eye position will blur the image. Game 1: You can experience gaze and saccades yourself. Look at one letter or short word on this page. As you keep your eyes fixed on that word, you are “gazing”. Any small shift in the position of your eye would blur the sharp edges of the word, so the eye needs to be maintained perfectly still. Now shift your gaze to another word on the page. What did you experience? The eye “jumps” quickly to the other word and then comes to another fixed position. That is the saccade. Notice that during the movement, you don’t register any of the words in between. The saccade is too fast. Also, notice that no matter how hard you try you can’t slowly and smoothly scan across the page picking up the in between words as you go. You might think that holding a gaze requires no muscle activity. After all the eye is just a floating ball in a socket, right? Actually, this is not the case. Look straight ahead and then shift your gaze to something to the left, keeping your head still. This produces a saccade by briefly activating extra-ocular muscles on the left of each eye (Figure 1). The leftward motion has a secondary consequence however, because it stretches the opposing tissues (the right extra-ocular muscles and ligaments) like a rubber band, which act to pull the eye back toward centre (Figure 2). Thus, to maintain your gaze to the left, it’s necessary to counteract this elastic rightward pull by activating the

A The argument set out here for why eye movement and posture are distinct is taken from Shadmehr (2016)1 where it is described in detail. B There are other types of eye control besides saccades and gaze holding like the vestibular ocular reflex and smooth pursuit which are omitted here for simplicity.

Figure 1. The extra-ocular muscles that move the eye. We will focus on the left and right opposing pair.

DRAFT – PLEASE DO NOT DISTRIBUTE same left extra-ocular muscles again, but far less vigorously and in a sustained manner - for as long as you sustain the position.

Figure 2. A large brief activation of the lateral rectus moves the eye and stretches the opposing muscle. A sustained and smaller activation of the same muscle is then required to oppose the stretch and hold the eye in place.

Tonic vs phasic modes To summarise, eye muscles have two different modes of activation: 1) the large transient bursts that generate saccades and 2) the continuous, low-level activity required to sustain posture. These two modes are called phasic and tonic respectively. These two words will be central to our discussion of posture and movement in general so it is worth the time to become familiar with them. Neural and muscular activity has long been observed to fall into one of these two different modes. Phasic activity is short-lived and rapidly changing while tonic is persistent and stable or slowly changing. For example, quick, planned body movements are typically phasic while postural activity is tonic. A fleeting thought might be phasic while a sustained emotional state like a state of calm might be more tonic. In eye muscles the difference is particularly stark with the extremely fast movements, the saccades, being generated by phasic activity and the very stable gaze by tonic. Eye circuitry For us, the important question about the eye is this: are the phasic and tonic activations of eye muscles controlled by one unified circuit or by two different circuits working together in parallel? In other words, is moving (saccades) and holding (gaze) controlled by one system or two? Believe it or not, the answer to this question will give us a tantalizing insight into what we may be doing as teachers.

DRAFT – PLEASE DO NOT DISTRIBUTE The brain circuitry for controlling the eye is located in specific clumps of neurons, called nuclei, in the brainstem. Researchers can measure which of these clumps are active as the eye alternates between saccades and gaze. It turns out that distinct areas “light up”, some corresponding to the control of saccades and others to the control of gaze. In other words, the experimental evidence demonstrates that saccades and gaze-holding are controlled separately, by specialized, relatively independent sub-circuits. These sub-circuits work hand in hand to control movement and posture (Figure 3).

Figure 3. A schematic of the eye control system. A saccade command specifies the rotation needed to move the eye to a target. The Saccade Centre in the brainstem translates this into a phasic burst of appropriate magnitude and duration that activates motor neurons that generate the saccade. A separate brainstem region, the Hold Centre, produces the sustained, tonic activity necessary to hold gaze once the saccade ends. Note that the Hold Centre and Saccade Centre are connected in parallel, with separate pathways to activate motor neurons and muscle.

The saccade sub-circuit consists of one nucleus that carries information about a target, namely how far the eye needs to rotate from its current position to reach it. Other nuclei then translate this into the phasic burst command that activates muscles to move the eye1. The parameters of the burst, its size and duration, are critical as they determine the saccade’s velocity and rotation respectively. If all goes to plan this lands the eye on target. Gaze is sustained by a different set of posture-related nuclei that generate the sustained tonic command necessary to fixate the eye. Remember that the muscles and ligaments that move the eye are also elastic. This means that to hold a gaze position, say to the left, the fixation command to the left turning muscles must perfectly counterbalance the opposing re-centering forces from the stretched tissues on the right. Any mismatch will cause the eyes to drift and blur vision. The further the gaze is from the center of the eye socket, the more tension is required to counteract the opposing stretched tissue. One might suppose that the eye’s posture sub-circuit is a simple reflex loop that senses when gaze drifts and corrects it reflexively. This is not the case. Instead, the posture sub-circuit generates the tonic command centrally, in the brain, without sensing eye position. In fact, even complete disconnection of the stretch receptors doesn’t affect gaze holding at

DRAFT – PLEASE DO NOT DISTRIBUTE all2. Instead, the brain monitors the saccade movement command to predict the next gaze location a priori and from this determines the size of the tonic hold command1. In this sense, one part of the brain watches what another is up to in order to be ready to meet the eventual postural demands of the movement. This evokes a quote by Sherrington that “posture follows movement like a shadow”C. For the eyes, the posture sub-circuit monitors movement plans to prepare the subsequent postural command, and then takes over. Because the saccade and gaze circuits are separate, the brain can tune each independently. This is important to understand when thinking about posture and movement in the broader sense. If the eyes regularly drift during gaze then the brain modifies the posture circuit without affecting the saccade command, and vice versa. A detailed explanation for why the the eye’s posture and movement circuits are considered distinct can be found in the Appendix. In summary, we see that the eye’s movement and posture are controlled by two separately tuned parallel systems that are concerned with different things – one with the burst parameters to move the eye and the other with an ongoing command to oppose disturbing forces.

BODY POSTURE The musculoskeletal system consists of numerous interlinked body segments and a network of muscles and ligaments that is vastly more mechanically complex than the eye. While much is known about the brain circuits that control the body, it is currently not understood as well as eye control, where the precise network and function of contributing brain regions is clear. Even the question of whether body posture and movement are controlled by distinct brain circuits is not certain. Nevertheless, from a task level perspective, we can think of movement as changing position and posture maintaining it. Are these tasks really controlled separately by two parallel systems, like the eye, or together, like an old fashioned hierarchical robot? Why might the answer to this be relevant for the AT? The rest of this article lays some of the groundwork for answering these questions.

Mechanical aspects of body posture Like the eye, holding a position in the body is a balancing act. Postural muscles need to match the force demands of the position. In the eye, this was a relatively simple affair, with right hand muscles matching left hand stretch and vice versa. In the body, distorting forces are more numerous and complex. We can divide these forces into internal forces - the mechanical stretches and pulls that arise inside the body - and external forces - gravity and contact forces that push and pull us from outside.

C Sherrington, C. (1906). The Integrative Action of the Nervous System. New York, NY: Charles Scribner's Sons.

DRAFT – PLEASE DO NOT DISTRIBUTE Internal forces The body, like the eye, generates internal forces from stretched muscles and ligaments that pull joints towards a central position. For example, turning your head to the left stretches ligaments and antagonistic muscles which pull back to the right. Thus, keeping your head left requires tonic activation of left-acting neck muscles to prevent the stretched tissues from returning your head toward centre (Figure 4). Turning your head further left increases the opposing pulls, requiring even greater muscle tone in left-acting muscles to maintain the position. This problem of matching internal stretches is exactly analogous to the problem we discussed of holding the eye stable.

Figure 4. Activity in left and right splenius capitis muscles before, during and after a head movement. The upper trace shows the head position as it moves from right to left. Before the movement, when the head is held to the right, tonic activity is seen in the right splenius to counteract stretched left-pulling tissues. A phasic burst in the left splenius moves the head to the left, which is now followed by tonic activity in the left splenius to counteract stretched right-pulling tissue and maintain posture. From Bizzi et al.3

External forces The outside world also exerts forces on the body that the motor system must contend with. These forces, which don’t play a significant role in the eye, are often larger than internal forces for the biomechanics of whole body movement. This greatly complexifies the problem of body posture and movement. There are two categories of external forces: gravity and contact forces. The postural system must counteract each of these, as well as internal forces, to maintain body position. Gravity Gravity pulls all body segments towards the earth. When standing, gravity acts to flex the hips, knees and spine, which if unopposed would cause postural collapse. Thus, postural

DRAFT – PLEASE DO NOT DISTRIBUTE tone in hip, knee and back extensors is necessary to keep the body erect. You’ve probably had the experience of your head dropping forward if you doze off on a plane. This occurs because sleep decreases postural tone allowing gravity to collapse the neck. The position of your body determines how gravity affects it. For example, standing in a monkey or tipping forward in a chair requires a very different postural response to gravity than standing or sitting up straight. Game 2: You can feel this postural activity adjust to gravity with a little experiment. Sit in a chair, place your hand on your low back and lean forward at the hips keeping your spine straight. Notice what happens to the tension of your back muscles as you lean forward. Your brain increases the activity of back extensors to counteract gravity’s increasing spinal torques.D Such interrelationships between gravity and body position hold true for habitual posture as well. For instance, habitually holding one’s head forward in front of the body requires increased postural tone in neck extensors. Contact forces Unlike gravity, contact forces are transmitted to body segments through physical contact with an object or the floorE. Pulling on a door handle, being bumped by someone and holding a cup of coffee all impart forces that disrupt posture. Contact forces are less predictable than both gravity and elastic forces from stretching tissues. They can also be transient — when walking a dog the tug on the leash can be quite brief. The postural system counteracts contact forces by activating muscles that oppose the disturbance, which must match the magnitude, direction, and timing of the external force. Returning to the dog-walking example, the short tug on the leash will be ideally opposed by a brief postural response. Disruptive forces also cause global effects. The tug on your hand from the dog leash sets up a chain of cascading forces: the leash pulls on the hand, which pulls on the forearm, which pulls on the upper arm, then shoulder, spine, legs and feet. To maintain posture throughout the brain must counteract forces across all of these joints. Game 3: You can experiment now with the whole-body nature of a postural response by pushing and pulling an immovable object like a closed door handle or a countertop while keeping yourself vertical and still (Figure 5). Notice the activity across the back and legs that occurs to maintain your posture as you apply forces with your hands. D This occurs because of gravity’s increased leverage on the spine when inclined. E To see the difference, consider jumping off of the floor and landing. The contact force of the floor only acts until your feet lift off the floor, while gravity pulls you down throughout your entire flight.

DRAFT – PLEASE DO NOT DISTRIBUTE

Figure 5. Push and pull while keeping vertically aligned and notice activity in front and back of body

Passive stability To some extent, the body mechanically counteracts both external and internal forces passively, in other words without immediate action from the nervous system. Due to their inherent stiffness, the bones themselves hold their shape fairly well against the compressional forces, such as from gravity. Passive stabilisation of the skeletal shape is also offered by the elastic resistance to stretch in soft tissues throughout the body. For example, picture a perfectly aligned stack of vertebra and imagine what would happen if they were bumped. They would, of course, be very vulnerable to being knocked over. Now add ligaments to your imaginary spine and repeat the experiment. The “ligamentous” spine is harder to knock over as the elasticity of ligaments resists vertebral motion and helps keep the spine upright. This only works for small disturbances, however, because the ligaments aren’t very stiff. In fact, the ligamentous spine buckles if you put roughly the weight of the head on top.4 Muscles enhance passive stability. Picture the same spine again but now add obliquely oriented muscles connecting the spine to the ribs and pelvis. These muscles act like guy wires on a radio mast. As with ligaments, stretching these muscles resists spinal motion and therefore imparts additional passive stability. Unlike ligaments, muscular tension can be changed by adjusting activity, so that toning them up increases the tension in the “guy wires” and therefore the passive stability of the spine.F Passive stability has the advantage that it imparts a basic level of stability that is always there and doesn’t involve the brain responding. However, it isn’t sufficient to counteract large disturbances to postureG, such as leaning forward at the hips, where back extensors are activated more strongly as you lean forward. Muscular redundancy

F The terminology is confusing here, but passive forces come only from stretch and not from increased activity in response to a disturbance. Even though the brain is activating muscles tonically, it is passive in that the brain doesn’t produce a response to the disturbance. G unless you are co-contracting intensely

DRAFT – PLEASE DO NOT DISTRIBUTE In the eye, only one muscle pulls in each direction. To hold a gaze to the left requires specific amount of tension in the left extra-ocular muscle. By contrast, the body’s many layers of overlapping muscles offer numerous different ways to maintain the same position. For instance, for a given position the trunk can be supported against gravity by deep or superficial muscles, medial or lateral muscles, a laterally asymmetric distribution and so on. This means that position alone does not determine the underlying distribution of support. While observing body position give us clues to what is going on, for example from a head far forward of the spine or a shortened neck, it doesn’t tell us the specific underlying pattern of support from position alone.

Neurological Control of Body Posture Postural tone “The greatest obstacle to discovery is not ignorance—it is the illusion of knowledge” — Daniel Boorstin For counteracting sustained forces like gravity, muscles are activated tonically to produce the appropriate amount of tension. Such activity, when it is produced automatically and unconsciously, is referred to as muscle toneH. For example the subconscious neck activity when sitting that we discussed, whose absence allows your head to fall forward when you fall asleep, is muscle tone. By contrast, the tonic activity when you voluntarily clench your fist is not considered to be tone because it is not automatic and subconscious. Sometimes the term postural tone is used instead when muscle tone plays a postural role such as opposing gravity. Even though muscle tone was one of the first motor phenomena to be studied (e.g. Sherrington 19245) it is still not well understood. There are both technical and historical reasons for this lack of understanding. Technically, muscle tone is difficult to quantify – it occurs across many muscles and is very slight. Historically, research into muscle tone has been impeded by the fallacy that it’s generated by the stretch reflex. Unfortunately this incorrect stretch reflex model has vastly oversimplified some peoples’ concepts of tone and postural support. Myth of the stretch reflex basis of tone

H This is a controversial term which has different meanings. For instance, in physiotherapy muscle tone is often defined as abnormal velocity dependent stretch reflexes caused by neurological damage. In general, confusion relates to the misunderstanding that muscle tone in healthy people is generated by the stretch reflex. Some fields have abandoned the term, however tonic activity in muscles is a real phenomenon, and the term muscle tone is current in the research literature.

DRAFT – PLEASE DO NOT DISTRIBUTE Many medical books still show stretch reflex diagrams as an explanation for how muscle tone is generated (Figure 6). This model is fundamentally and incorrigibly flawed. The main problem with this perspective is that the stretch reflex doesn’t generate ongoing activity – it is phasicI. Hitting the patellar tendon with a rubber doctor’s hammer activates a stretch reflex pathway resulting in a “knee jerk”. But this rapid jerk response to a fast muscle pull contrasts markedly with muscle tone.

Figure 6. A typical incorrect diagram for how muscle tone is generated.

Game 4 Go ahead and try and to alter muscle tone in your quadriceps with this method so that your knee elevates by a few degrees and stays there. Perhaps experiment by gently hitting your tendon, pressing slowly or even pushing on it persistently. You will of course find that this doesn’t work. Think back to the eye. Recall that disconnection of eye muscle stretch receptors had no effect on tonic activity during gaze fixation. In fact, in the body as well, stretch reflexes don’t generate muscle tone.J In the absence of brain damage, high velocity stretch (above 100 degrees per second) is required to elicit a stretch reflex; below that it produces no response at all. As tone is necessary in static postures without rapid stretching, it is difficult to explain how this stretch reflex generates tone. It is also notable that muscle tone is present in patients without functional stretch receptors.7 Central regulation of tone

I A persistent, tonic stretch reflex has been described with brain damage, however it cannot be reliably elicited in healthy individuals and its neural pathways are not known. It is almost certainly a complex response possibly involving higher brain areas. The term tonic stretch reflex is not currently used. J For a review of arguments against the stretch reflex model of tone see Davidoff6.

DRAFT – PLEASE DO NOT DISTRIBUTE The historical stretch reflex explanation of tone neglects how the brain itself can generate and shape muscle tone. The real story is much more complex and at times mysterious. While there are some brain regions that generate tonic activity, such as the brainstem, most of the brain’s neurons are phasic. For example, examine the output of the motor cortex, the primary high-level motor output station, as subjects both moved a stiff joystick to an extreme position and then held it there against a force (Figure 7).8 Neurons in the spinal cord are active during both move and hold periods. Notably, they are tonically active during the hold period to produce the tone necessary to maintain the joystick’s position. Contrast that with neurons in the motor cortex which were only active phasically, i.e. transiently during the reach but silent during the the hold period. This begs the question, where does the sustained signal come from that’s driving the spinal neurons tonically?

Figure 7A. Firing from a cortical neuron and a spinal neuron while reaching to a target. The cortical neuron fires two bursts, while the spinal neuron fires tonically. The first burst in the cortical occurred when the instruction to move was given and the second was when the arm actually moved. From Shalit et al 2012.

Figure 7B. Averaged activity from cortical neurons and spinal neurons during a reach and hold action. Note how the red cortical trace increases during the movement then decreases when a position is held. In contrast, the spinal neurons stay active during the hold posture. From Shalit et al 2012.

There is evidence that neurons in the brainstem provide this sustained signal that generates muscle tone. Even brief stimulation to one such region causes sustained neck postures. Damage to the same region renders one unable to maintain head position. Similar brainstem regions exist for the body which, when briefly stimulated, cause sustained changes in bodily muscle tone, for example increasing it, decreasing it or changing its distribution across muscles.9 The current hypothesis is that such brainstem regions act as a “neural integrator” that converts various phasic signals into the sustained drive that underlies muscle tone. Here the word “integrator” is used in the sense of calculus, when one movement signal that may rapidly vary up and down is converted through a simple mathematical calculation into a slowly increasing signal that persists even when the movement stops (Figure 8). This is thought to work analogously to the eye hold circuit, although the body circuit is thought to be more complex, integrating information from more sources, such as sensory feedback,

DRAFT – PLEASE DO NOT DISTRIBUTE presumably to tailor muscle tone to match external loads.K Damage to this neural integrator or its various inputs is hypothesised to cause dystonia, such as torticollis.

Figure 8. The brainstem neural integrator hypothesis 1,10,11. High level phasic signals of various forms all project to the neural integrator, which converts these into a sustained tonic output that projects through the spinal cord to muscle to produce muscle tone.

One recently documented case of note is a patient with idiopathic camptocormia, or “bent spine” syndrome. There are various causes of camptocormia, but this woman in particular had no other deficits but the inability to maintain trunk posture12. Over a minute or so her spinal posture went from vertical to bent way over - 60° forward (Figure 9). While her particular neural defect was unknown, her behaviour suggests a specific deficit to the brainstem neural integrator or related regions. It is striking that she can “stand up straight” when voluntarily asked, but when distracted her posture collapses way forward because the muscle tone in her back extensors isn’t properly regulated. The volitional standing up straight isn’t considered muscle tone because it isn’t automatic, and is clearly regulated by a separate brain region as it remains normal.

K While torticollis is considered a “dystonia”, prevailing hypotheses do not generally consider tone as there is such poor understanding of its physiology.

DRAFT – PLEASE DO NOT DISTRIBUTE

Figure 9. A patient with camptocormia who cannot maintain her trunk posture without attention. Right shows how her trunk gradually inclines over several minutes.

Tone and the Alexander Technique Let’s now consider how - or whether - the AT affects muscle tone. Anecdotally it would seem to as it changes how you carry yourself, and therefore presumably the distribution of tone. Consider specific anecdotal evidence from AT lessons. Typically, a teacher might use a combination of hands on and verbal cueing to create subtle shifts in the persistent patterns of tension as a student is standing, sitting, or lying down. These patterns are not directly consciously controlled (like the clenched fist in our previous example) in the sense that the student may be unaware of the particular pattern until it changed. The shifts occur slowly, with repetition over the entire period of the lesson rather than suddenly. The shifts may result in changes in the student’s experience of gravity (they feel lighter) and freedom to move (ease) and may persist for minutes, hours, or even days. All of this evidence, the subconscious nature, the persistence, the slow rate of change, and the relation to gravity, suggests that this aspect of an AT lesson involves shifts in postural tone. Game 5 Consider moments in a recent lesson that you gave or received when you experienced a change in tone. What distinguishes these shifts in tone from other moments in the lesson? Happily there are also several lines of more solid scientific experimental evidence. The first involved assessing tone on a novel device called Twister, which measures the resistance to extremely slow twisting of the neck, trunk and hips while standing. The measured resistance quantifies tone as it quantifies the internal muscular forces used to counteract gravity. It’s very much the same principle as turning an upright student’s head using free wrists and gauging the resistance. The data showed that Alexander training and AT teachers were

DRAFT – PLEASE DO NOT DISTRIBUTE associated with lower resistance in all regions tested.13 At the same time Alexander teachers showed a distinctly different regulation of tonic activity, with more adaptability in trunk muscle tone during the motion than controls, so that the distribution of support readily changed to accommodate shifts in position. Further analysis found that this adaptable tone is one major cause of the decreased stiffness. Other causes, such as alignment or the specific distribution of tone may also have played a role. There is some research suggesting the AT also changes the distribution of postural activity from superficial to deeper muscles in the neck. Frank Jones found that, for a small group of subjects, AT guidance decreased activity of the sternocleidomastoid in sitting.14 A more recent and larger study also found a decrease in sternocleidomastoid to opposing external loading from AT lessons.15 For the protocol used, it’s known that reduced sternocleidomastoid activity is associated with greater activity in deep neck muscles, suggesting that the AT shifts activity from superficial to deep muscles. Two additional points of note regarding tone. Muscle tone is highly individual – Twister found it to be extremely consistent within someone over months but very different across people.13,16 This seems in accord with habitual differences between people observed in AT practice. Secondly, muscle tone is very slow to adapt. Twister and other studies have found it can take tens of seconds to change distribution, which may be related to the slow changes that occur when putting hands on in lessons. These slow, lasting changes contrast markedly with the immediate change produced by changing posture volitionally.

MOVEMENT Movements in the body are far more diverse than those in the eye (e.g. saccades). Bodily movement is often divided into three broad classes: automatic, voluntary and rhythmic actions. Examples of these are sneezing, throwing and running, respectively. Despite this diversity, movements all share one thing in common – they result from an imbalance of forces that cause the motion.L In this sense, movements are the opposite of maintaining a posture, which requires the equilibration of forces.

L Technically forces cause accelerations that start, stop and shape movements. This is a consequence of Isaac Newton’s second law, force = mass x acceleration.

DRAFT – PLEASE DO NOT DISTRIBUTE Moving part of the body, say your arm, along a trajectory to a target results from discrete sequence of events throughout time. Muscles are activated to accelerate the arm, imparting it with a velocity, for instance deltoid and triceps might be activated to raise and extend the arm. Then some time later antagonist muscles are activated to brake the motion, for instance biceps might be used. While the movement may feel smooth, the muscles are activated in phasic bursts. The timing and size of these bursts determines the movement trajectory. For example, the arm will overshoot if the braking burst is late. Movement coordination refers to the act of orchestrating the timing and size of the various phasic bursts across muscles so that the action achieves the desired trajectory. In an action like walking this is a non-trivial task, requiring the precise timings across many muscles on both sides of the body (Figure 10). Game 6: Move your arm back and forth as if reaching for a cup of coffee, with the arm suspended in the air. Take a moment to consider the complexity of the movement – multiple joints need to be accelerated and decelerated with just the right timing – and nevertheless the hand ends up where you wanted it to go even if you do the movement very quickly. Contrast this with the postural elements – the hand suspended in the air – which require persistent signaling to the muscles to meet gravity. Neurological basis of movement While a survey of how the brain controls movement is beyond the scope of this article, suffice it to say that the main task for the brain is to coordinate the timings of the various phasic bursts to bring about the desired action. The phasic bursts can be generated at high levels of the nervous system, like playing a piano, or low levels like coughing, breathing or the basic walking pattern. The involvement of higher brain levels like the motor cortex allows a substantial degree of flexibility to shape the pattern at will, in contrast to the stereotypy produced by low levels such as the brainstem or spinal cord.

Figure 10. Muscle activity during walking. Note how that even though it seems continuous, muscles are activated in bursts. From Paci et al.17

DRAFT – PLEASE DO NOT DISTRIBUTE While a robot coordinates movement by sensing and shaping what’s happening in real time as it unfolds, the movement system’s ability to use feedback during rapid movement is limited because neurons transmit information far too slowly. For instance, if your brain tried to sense when your arm nears a target to trigger the antagonist braking burst, it would be too late to stop the arm in time. When the information eventually comes in it is too late. Game 7: Try moving your finger rapidly to a very small target, for example putting your finger down on the corner of a table or the tip of a pencil point. Notice that while the rough movement happens very quickly, the final few centimetres happen very slowly. This is because the last bit of precise movement requires using feedback from sensory organs to bring the finger to land, and this feedback is much too slow for rapid movement. As a result of slow feedback, the brain relies heavily on advanced planning to make most movements. For example, the timings of the various bursts of leg muscles in steady walking are produced by a central pattern generator in the spinal cord, which can produce the pattern in the complete absence of sensory feedback. Voluntary movements are planned more flexibly by the cortex, but the details of the movement – from the trajectory to the velocity to the grasp you will use – are planed in advance.

ARE POSTURE AND MOVEMENT CONTROL DISTINCT? Before examining the evidence, let’s consider what it might mean for the AT if movement and posture are controlled separately by distinct circuits. This would suggest that posture and movement can affect and even interfere with one another. Think again about the anecdotal evidence given above for how tone changes in AT lessons. A table work session is largely a shift in tone, by definition. The hands on work and verbal cueing of the teacher combined with the thinking of the student shifts the largely subconscious or semiconscious patterns of tension throughout the student’s body. These shifts persist after the table work, changing the posture of the student and overall feeling of support against gravity. Furthermore the table session may have changed the way the student moves – even without specific instructions regarding movement planning. In other words, a stiff or poorly supported student who moves jerkily before the session may end up moving with better support and more smoothness after the session. This suggests posture (the patterns of muscle tone) affects patterns of movement – for better or for worse. Currently there is not conclusive scientific evidence that posture and movement are controlled separately. But there is evidence in that direction, which when combined with our experiences as AT teachers makes for a compelling case. Recall for example the experiment described above where brainstem deactivation eliminated the ability to maintain neck posture. This elimination of postural control left movement unchanged. While it was possible to turn the head to a target, it returned to centre when trying to keep it there.18 Similarly, consider the camptocormia patient above who cannot maintain her trunk posture against gravity. Despite this deficit her movement was normal. Another

DRAFT – PLEASE DO NOT DISTRIBUTE relevant experiment played “tricks” on subjects by distorting their perception of arm position during reaching movements. 19 This caused the movement planning system to adapt. However, after adapting, when a position was held it drifted back to the original “untricked” location, suggesting that the movement system adapted independently to the postural system. Finally, consider a one year old child, who shows beautiful postural support but has still not learned the simplest of movement or balance patterns (Figure 11). While this dramatic difference in development circumstantially

suggests a degree of independence in the systems, tone is known to develop long before movement coordination and balance.

Movement and the AT

The question remains as to why the AT changes movement. We know it does, for instance take the sit-to-stand movement. AT teachers can rise from a chair with less intervertebral motion and a smoother lift-off than untrained subjects.20 The question is what causes this. To an outsider, the obvious possibility is that the AT changes movement planning, in other words directly specifies a different trajectory by altering the phasic bursts that underlie the movement. In other words standing up “AT style”. But this is not consistent with anecdotal descriptions of AT lessons, as teachers don’t typically cue students to “first do this, then do this, then do that…” to alter movement. The other possibility, of course, is that the AT affects the movement indirectly through changes to the postural system. We performed an experiment to examine whether AT changes to sit-to-stand resulted from movement planning.21 By slowing the chair rise down to a near glacial pace, we ensured that movement planning would be easy, like slowing down a piano piece for a beginner. If the differences were due to movement planning, then controls should be able to better mimic AT teachers’ coordination for such slow movements. Instead, we found the opposite was true – controls found slow movements harder and had a pronounced lurch at liftoff that they were unable to prevent. A biomechanical computer model suggests their lurch could be present to compensate for excess postural stiffness. Looking at chair work, it almost seems obvious that the AT affects movement through posture. A teacher makes adjustments to a seated student’s postural behaviour and then

Figure 11. A man balancing a baby on his hand

DRAFT – PLEASE DO NOT DISTRIBUTE both teacher and student observe how this change affects movement as the student leaves the chair. Game 8 Consider this discussion of AT lessons, posture and movement. Is it consistent with your experience? Does this model of posture and movement as parallel systems help you think about lessons? Or not? Conclusions As we have seen, despite the dictionary definition posture is not just about position. While the postural system aims to regulate position, it is a complex interplay of many things: internal and external forces, position, distribution of muscle tone and voluntary postural behaviour. While the control of postural tone is still poorly understood, it is not generated by the stretch reflex. Regions of the brainstem have a role in generating and tuning muscle tone, and there is substantial recent scientific interest and progress to understand these circuits. Experiments have demonstrated that the AT changes postural tone to make it more adaptive and shifts its distribution toward deeper spinal muscles. While it is too early to state conclusively, there is evidence that body posture and movement are regulated by distinct parallel circuits, like the eye. Finally, it appears that the AT may affect movement though the postural system.

Appendix The eye’s posture and movement circuits are considered distinct because the two circuits operate and adapt independently. This is supported by three different lines of evidence: 1) Each sub-circuit can be impaired separately. For example, damage to the movement nuclei causes saccades to miss their targets but doesn’t affect gaze fixation. On the other hand, damage to eye posture nuclei doesn’t affect saccades, but then the eye can’t hold gaze and drifts back to centre. 2) Saccades and gaze adapt independently. Saccades can be diminished in size by repeatedly moving target locations closer mid-saccade. This adaptation is specific to eye movement and doesn’t affect gaze-fixation. On the other hand, eye posture can also be tricked through simulating eye drift (by moving the whole visual scene) during periods of gaze fixation. This eventually causes the posture sub-circuit to adjust gaze holding tension, to eliminate the imposed drift, but saccades are unaffected. 3) Adaptation to eye movement and eye posture, such as just described, occurs via different circuits. The oculomotor vermus of the cerebellum adapts the former while the flocculus adapts the latter, indicating the brain treats eye movement and posture separately. Intuitively this makes sense as movement is concerned with changing a state (gaze position), while posture is concerned with keeping it the same.

DRAFT – PLEASE DO NOT DISTRIBUTE References 1 Shadmehr, R. Distinct neural circuits for control of movement vs. holding still. J

Neurophysiol 117, 1431-1460, doi:10.1152/jn.00840.2016 (2017). 2 Lewis, R. F., Zee, D. S., Hayman, M. R. & Tamargo, R. J. Oculomotor function in the

rhesus monkey after deafferentation of the extraocular muscles. Exp Brain Res 141, 349-358, doi:10.1007/s002210100876 (2001).

3 Bizzi, E., Kalil, R. E. & Tagliasco, V. Eye-head coordination in monkeys: evidence for centrally patterned organization. Science (New York, N.Y 173, 452-454, doi:10.1126/science.173.3995.452 (1971).

4 Lucas, D. B. & Bresler, B. Stability of the ligamentous spine. Report No. esr. 11 No. 40, (University of California, San Francisco, 1960).

5 Sherrington, C. & Liddell, E. G. T. Reflexes in Response to Stretch (Myotatic Reflexes). Royal Society of London Proceedings Series B 96, 212 (1924).

6 Davidoff, R. A. Skeletal muscle tone and the misunderstood stretch reflex. Neurology 42, 951-963 (1992).

7 Cole, J. Losing touch : a man without his body. First edition. edn, (Oxford University Press, 2016).

8 Shalit, U., Zinger, N., Joshua, M. & Prut, Y. Descending systems translate transient cortical commands into a sustained muscle activation signal. Cereb Cortex 22, 1904-1914, doi:10.1093/cercor/bhr267 (2012).

9 Takakusaki, K., Chiba, R., Nozu, T. & Okumura, T. Brainstem control of locomotion and muscle tone with special reference to the role of the mesopontine tegmentum and medullary reticulospinal systems. J Neural Transm (Vienna) 123, 695-729, doi:10.1007/s00702-015-1475-4 (2016).

10 Joshua, M. & Lisberger, S. G. A tale of two species: Neural integration in zebrafish and monkeys. Neuroscience 296, 80-91, doi:10.1016/j.neuroscience.2014.04.048 (2015).

11 Shaikh, A. G., Zee, D. S., Crawford, J. D. & Jinnah, H. A. Cervical dystonia: a neural integrator disorder. Brain 139, 2590-2599, doi:10.1093/brain/aww141 (2016).

12 St George, R. J., Gurfinkel, V. S., Kraakevik, J., Nutt, J. G. & Horak, F. B. Case Studies in Neuroscience: A dissociation of balance and posture demonstrated by camptocormia. J Neurophysiol 119, 33-38, doi:10.1152/jn.00582.2017 (2018).

13 Cacciatore, T. W., Gurfinkel, V. S., Horak, F. B., Cordo, P. J. & Ames, K. E. Increased dynamic regulation of postural tone through Alexander Technique training. Human Movement Sci 30, 74-89, doi:S0167-9457(10)00156-9 [pii].

14 Jones, F. P., Hanson, J. A. & Gray, F. E. Head balance and sitting posture II: the role of the sternomastoid muscle. Journal of Psychology 52, 363-367 (1961).

15 Becker, J. J., Copeland, S. L., Botterbusch, E. L. & Cohen, R. G. Preliminary evidence for feasibility, efficacy, and mechanisms of Alexander technique group classes for chronic neck pain. Complement Ther Med 39, 80-86, doi:10.1016/j.ctim.2018.05.012 (2018).

16 Gurfinkel, V. et al. Postural muscle tone in the body axis of healthy humans. J Neurophysiol 96, 2678-2687, doi:10.1152/jn.00406.2006 (2006).

DRAFT – PLEASE DO NOT DISTRIBUTE 17 Paci, D. et al. Subclinical abnormal EMG activation of the gastrocnemii during gait

analysis in restless legs syndrome: A preliminary report in 13 patients. Sleep Medicine 10, 312-316, doi:10.1016/j.sleep.2008.04.007 (2009).

18 Klier, E. M., Wang, H., Constantin, A. G. & Crawford, J. D. Midbrain control of three dimensional head orientation. Science (New York, N.Y 295, 1314-1316 (2002).

19 Ghez, C., Scheidt, R. & Heijink, H. Different learned coordinate frames for planning trajectories and final positions in reaching. J Neurophysiol 98, 3614-3626, doi:10.1152/jn.00652.2007 (2007).

20 Cacciatore, T. W., Gurfinkel, V. S., Horak, F. B. & Day, B. L. Prolonged weight-shift and altered spinal coordination during sit-to-stand in practitioners of the Alexander Technique. Gait Posture 34, 496-501, doi:10.1016/j.gaitpost.2011.06.026 (2011).

21 Cacciatore, T. W., Mian, O. S., Peters, A. & Day, B. L. Neuromechanical interference of posture on movement: evidence from Alexander technique teachers rising from a chair. J Neurophysiol 112, 719-729, doi:10.1152/jn.00617.2013 (2014).